Front vehicle trajectory generation method and module

The trajectory generated by perception and fitting functions solves the problem of car follow-up trajectory in the missing lane line and complex scenarios, and achieves the improvement of robustness and safety in low-speed assisted driving, ensuring the rapid response and smooth car follow-up of functions .

CN115683116BActive Publication Date: 2024-11-01LIANCHUANG AUTOMOBILE ELECTRONICS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211362270.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-02
Publication Date
2024-11-01
Estimated Expiration
2042-11-02

AI Technical Summary

Technical Problem

In the absence of lane lines and complex scenarios, it is difficult for existing low-speed assisted driving systems to quickly and accurately generate follow-up tracks, resulting in insufficient functional robustness and safety.

Method used

By perceiving the target information, judging new targets or historical targets, integrating or updating trajectory points, using the fitting function to form a vehicle driving trajectory, selecting parabolic trajectory and equal distribution trajectory points, and fitting through the least squares method with constraints, constructing The objective function is used to generate a smooth follow-up track.

Benefits of technology

It realizes rapid trajectory generation in unclear lane lines and complex scenarios, enhances the robustness and safety of assisted driving functions, reduces driving risks, and ensures the smoothness and functional response of the car follow-up process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115683116B_ABST
    Figure CN115683116B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for generating a preceding vehicle trajectory, including: sensing a target, judging whether it is a new target or a historical target; if it is a new target and there is a historical trajectory, integrating the new target into its historical trajectory; if it is a new target and there is no historical trajectory, calculating the designated trajectory of the new target through the position of the vehicle and the position of the preceding vehicle; if it is a historically valid target, updating the historically valid target information to the information at the last moment; if it is a historically invalid target, not processing it; arranging trajectory points for describing the preceding vehicle's historical trajectory, subsequent trajectory updates, and subsequent trajectory fitting on the designated trajectory; forming a following vehicle driving trajectory through a fitting function. The present invention can quickly generate a following vehicle trajectory, can achieve stability in the following vehicle process, and suppress the mutation phenomenon when switching between following the vehicle and following the line, and can enhance the robustness and safety of the assisted driving function in complex scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of automobiles, and in particular to a method for generating a preceding vehicle track in intelligent driving technology; and a preceding vehicle track generating module. Background Art

[0002] In recent years, with the rapid development of the intelligent driving industry, the demand for low-speed assisted driving functions in urban and rural road conditions has been increasing. However, the centering function based only on lane line information will be greatly reduced in robustness when encountering special scenarios such as unclear lane lines, intersections without guide lines, and lane lines covered by road vehicles, affecting the driving experience of drivers and passengers. The front vehicle trajectory planning algorithm can quickly plan path information that meets the needs of following the vehicle based on the front vehicle information input by the front camera, which is used to replace the lane line information required by the centering function to ensure the overall robustness requirements of the lateral function.

[0003] At the same time, the trajectory planning algorithm for the vehicle ahead takes into account the complex scenario of the constantly changing targets ahead during the following process, so as to ensure the comfort of following the vehicle while avoiding the risk of following the vehicle. Therefore, a stable trajectory planning algorithm that takes into account complex scenarios and can quickly generate the trajectory of the vehicle ahead is an indispensable component of low-speed assisted driving and has a very broad application prospect.

[0004] In the existing low-speed assisted driving function, trajectory generation mainly includes the following two design ideas:

[0005] 1. When encountering special conditions such as unclear lane lines and intersections without guide lines, the assisted driving function will give up control of the vehicle and instead remind the driver to take over, which greatly reduces the robustness of the function;

[0006] 2. The position information of the on-path vehicle (the closest vehicle in the vehicle's path) is input by the camera, and the least squares method is simply used to fit the trajectory of the on-path vehicle. However, the trajectory generation is slow, and behaviors such as ignoring the target switching of the front vehicle and overfitting will bring driving risks during the following process.

[0007] Therefore, a trajectory planning algorithm that improves functional robustness while ensuring functional comfort and safety is essential. Summary of the invention

[0008] A series of simplified concepts are introduced in the Summary of the Invention, which are all simplifications of the prior art in the field, and will be further described in detail in the Detailed Description of the Invention. The Summary of the Invention of the present invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0009] The technical problem to be solved by the present invention is to provide a method for generating a leading vehicle trajectory that can quickly and accurately generate a following vehicle requirement in the absence of lane lines and / or in complex scenarios.

[0010] And, a leading vehicle trajectory generation module that can quickly and accurately generate the following vehicle requirements in the absence of lane lines and / or complex scenarios.

[0011] In order to solve the above technical problems, the present invention provides a method for generating a preceding vehicle trajectory, comprising the following steps:

[0012] S1, perceive the target and judge whether it is a new target or a historical target;

[0013] S2: If it is a new target and there is a historical trajectory, the new target is integrated into its historical trajectory; if it is a new target and there is no historical trajectory, the new target specified trajectory is calculated by the position of the vehicle and the position of the preceding vehicle;

[0014] If it is a historically valid target, the current historically valid target information is updated to the information of the previous moment; further explanation: the trajectory is fitted by the points composed of the vehicle positions, so when the target position is updated and is a historical target, it is updated to the information of the previous moment for subsequent fitting; if it is a historically invalid target, it is not processed;

[0015] S3, by arranging trajectory points on the designated trajectory for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates, and subsequent trajectory fitting;

[0016] S4, forming a vehicle-following driving trajectory by fitting the function.

[0017] Optionally, the preceding vehicle trajectory generation method is further improved, wherein the designated trajectory is a parabola.

[0018] Optionally, the preceding vehicle trajectory generation method is further improved so that the trajectory points are arranged in an equal-division manner.

[0019] Optionally, the preceding vehicle trajectory generation method is further improved, and step S4 includes the following sub-steps:

[0020] S4.1, construct objective function: objective function = ∑ (observed value - theoretical value) 2 ;

[0021] The observed value is the target position information obtained through steps S1 and S2, the theoretical value is the assumed fitting function, and the objective function is the loss function;

[0022] S4.2, obtaining the fitting function corresponding to the minimum value of the objective function;

[0023] S4.3, forming a formula to express the vehicle-following driving trajectory;

[0024]

[0025]

[0026] i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point;

[0027] j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight. .

[0028] In order to solve the above technical problems, the present invention provides a preceding vehicle trajectory generation module, comprising:

[0029] A receiving unit, wherein the self-sensing system receives target information, wherein the target information includes location information, index information and validity;

[0030] Judgment unit, judging whether it is a new target or a historical target;

[0031] The trajectory generation unit, if it is a new target and there is no historical trajectory, merges the new target into its historical trajectory; if it is a new target and there is no historical trajectory, calculates the new target specified trajectory through the position of the vehicle and the position of the preceding vehicle;

[0032] If it is a historically valid target, the current historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed;

[0033] The fitting unit arranges trajectory points for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates and subsequent trajectory fitting on the specified trajectory, and forms a following vehicle driving trajectory through a fitting function.

[0034] Optionally, the preceding vehicle trajectory generation module is further improved, and the designated trajectory is a parabola.

[0035] Optionally, the preceding vehicle trajectory generating module is further improved so that the trajectory points are arranged in an equally divided manner.

[0036] Optionally, the preceding vehicle trajectory generation module is further improved, and the fitting unit forms the following vehicle driving trajectory in the following manner:

[0037] Construct objective function: objective function = ∑ (observed value - theoretical value) 2 ;

[0038] The observed value is the target position information output by the trajectory generation unit, the theoretical value is the assumed fitting function, and the objective function is the loss function;

[0039] Obtain the fitting function that makes the objective function obtain the minimum value;

[0040] Form a formula to express the vehicle-following driving trajectory;

[0041]

[0042]

[0043] i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point;

[0044] j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight.

[0045] In order to improve the stability of the assisted driving function in unclear and / or complex scenarios (such as intersections), the present invention is based on the target information (including position, index and validity, etc.) input by the environmental perception sensor (such as a camera), and through new targets and their validity judgment, specifying the front vehicle trajectory, fitting functions and other technical means to quickly generate a driving trajectory that meets the centering control requirements, which is used for the technical problem of the centering driving of the vehicle.

[0046] The present invention can quickly generate a following vehicle trajectory, can achieve rapid function response, and timely compensate for the centering performance in special scenarios; through constrained least squares curve fitting, a smooth following vehicle trajectory that ignores the curvature change rate is generated, which can achieve stability in the following process and suppress the sudden change phenomenon when switching between following the vehicle and following the line; detailed information screening logic can realize the switching of following vehicle targets, eliminate the driving risks caused by cutting in and out of the conditions during following the vehicle, and can enhance the robustness and safety of the assisted driving function in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings of the present invention are intended to illustrate the general characteristics of the methods, structures and / or materials used in specific exemplary embodiments of the present invention, and to supplement the description in the specification. However, the drawings of the present invention are schematic diagrams not drawn to scale, and thus may not accurately reflect the precise structure or performance characteristics of any given embodiment, and the drawings of the present invention should not be interpreted as limiting or restricting the range of values ​​or properties covered by the exemplary embodiments of the present invention. The present invention is further described in detail below in conjunction with the drawings and specific embodiments:

[0048] Figure 1 It is a schematic diagram of the target screening process of the present invention.

[0049] Figure 2 It is a schematic diagram of the principle of the present invention. DETAILED DESCRIPTION

[0050] The following is an explanation of the embodiments of the present invention by specific specific embodiments, and those skilled in the art can fully understand other advantages and technical effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through different specific embodiments, and the details in this specification can also be applied based on different viewpoints, and various modifications or changes can be made without deviating from the general design concept of the invention. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. The following exemplary embodiments of the present invention can be implemented in a variety of different forms and should not be interpreted as being limited to the specific embodiments described herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical solutions of these exemplary specific embodiments to those skilled in the art. It should be understood that when an element is referred to as "connected" or "coupled" to another element, the element can be directly connected or coupled to another element, or there can be an intermediate element. The difference is that when an element is referred to as "directly connected" or "directly coupled" to another element, there is no intermediate element.

[0051] First embodiment;

[0052] The present invention provides a method for generating a preceding vehicle trajectory, comprising the following steps:

[0053] S1, perceive the target and judge whether it is a new target or a historical target;

[0054] S2: If it is a new target and there is a historical trajectory, the new target is integrated into its historical trajectory; if it is a new target and there is no historical trajectory, the new target specified trajectory is calculated by the position of the vehicle and the position of the preceding vehicle;

[0055] If it is a historically valid target, the current historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed;

[0056] S3, by arranging trajectory points on the designated trajectory for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates, and subsequent trajectory fitting;

[0057] S4, forming a vehicle-following driving trajectory by fitting the function.

[0058] Second embodiment;

[0059] The present invention provides a method for generating a preceding vehicle trajectory, comprising the following steps:

[0060] S1, perceive the target and judge whether it is a new target or a historical target;

[0061] S2, if it is a new target and there is a historical trajectory (cut-in scene), sort out the historical trajectory that exceeds the new target position, that is, merge the new target into its historical trajectory;

[0062] If it is a new target and there is no historical trajectory (initial target), the new target specified trajectory is calculated by the position of the vehicle and the position of the preceding vehicle. In this embodiment, it is assumed that the current trajectory is a parabola;

[0063] If it is a historically valid target, the current historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed;

[0064] In order to improve the robustness of the function in complex scenarios (unclear lane lines, intersections, etc.), it is very necessary to ensure that the trajectory information of the preceding vehicle can be quickly generated when the lane line information is invalid. However, in some special scenarios (for example, the preceding vehicle starts at a standstill at an intersection), there is no historical trajectory of the preceding vehicle, and it is impossible to fit a preceding vehicle trajectory that meets the requirements according to the existing technology. Therefore, in this scenario, this embodiment assumes that the preceding vehicle trajectory is a parabola, and the trajectory of the preceding vehicle can be calculated through the position of the vehicle and the preceding vehicle for fast following;

[0065] S3, by arranging trajectory points on the designated trajectory for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates, and subsequent trajectory fitting;

[0066] The trajectory points are arranged in an appropriate amount by equally dividing the parabola. For example, the parabola is equally divided into 30 points to generate the trajectory point set of the preceding vehicle, and the count is 30;

[0067] S4, forming a vehicle-following driving trajectory by fitting the function.

[0068] Determine whether the number of target trajectory points meets the requirement for further target trajectory fitting. If the condition is not met, reset the trajectory, which is equivalent to clearing the trajectory to 0.

[0069] In order to avoid fitting a smoother curve and meet the comfort requirements of following the vehicle, the constrained least square method is introduced to calculate the polynomial coefficients and error square of the trajectory curve of the leading vehicle under the constraint conditions;

[0070] The confidence information of the target trajectory generated by the mean square error calculation is used as the basis for judging the reliability of the subsequent trajectory. The specific process is as follows:

[0071] Use the constrained least squares method to fit a quadratic polynomial curve for following the vehicle (through testing, it was found that the stability of the cubic polynomial with the curvature change rate term is worse). The least squares method is a mathematical tool that is widely used in many disciplines such as error estimation, uncertainty, system identification, prediction, and forecasting data processing. The objective function is constructed in the following form:

[0072] Objective function = ∑(observed value - theoretical value) 2

[0073] The observed value is the target position information obtained through steps S1 and S2, the theoretical value is the assumed fitting function, and the objective function is the loss function;

[0074] The goal is to obtain the fitting function that minimizes the loss function. Considering the comfort of the following process, ignoring the steering wheel shaking caused by the change in curvature rate, the quadratic polynomial is selected as the fitting function, and in order to avoid overfitting, the L2 regularization term is introduced to reduce overfitting (avoid the occurrence of certain extreme working conditions), and finally the following driving trajectory expression formula is formed;

[0075]

[0076]

[0077] i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point;

[0078] j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight.

[0079] Third embodiment;

[0080] The present invention provides a preceding vehicle trajectory generation module, which can be implemented based on the hardware and computer programming technology means of the prior art, including:

[0081] A receiving unit, wherein the self-sensing system receives target information, wherein the target information includes location information, index information and validity;

[0082] Judgment unit, judging whether it is a new target or a historical target;

[0083] The trajectory generation unit, if it is a new target and there is no historical trajectory, merges the new target into its historical trajectory; if it is a new target and there is no historical trajectory, calculates the new target specified trajectory through the position of the vehicle and the position of the preceding vehicle;

[0084] If it is a historically valid target, the historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed;

[0085] The fitting unit arranges trajectory points for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates and subsequent trajectory fitting on the specified trajectory, and forms a following vehicle driving trajectory through a fitting function.

[0086] Third embodiment;

[0087] The present invention provides a preceding vehicle trajectory generation module, which can be implemented based on the hardware and computer programming technology means of the prior art, including:

[0088] A receiving unit, wherein the self-sensing system receives target information, wherein the target information includes location information, index information and validity;

[0089] Judgment unit, judging whether it is a new target or a historical target;

[0090] The trajectory generation unit, if it is a new target and there is no historical trajectory, merges the new target into its historical trajectory; if it is a new target and there is no historical trajectory, calculates the new target's specified parabolic trajectory through the position of the vehicle and the position of the preceding vehicle;

[0091] If it is a historically valid target, the historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed;

[0092] A fitting unit, which forms a following vehicle driving trajectory by equally arranging trajectory points for describing the historical trajectory of the preceding vehicle, subsequent trajectory update and subsequent trajectory fitting on the designated trajectory through a fitting function;

[0093] The fitting unit forms the following vehicle trajectory in the following way:

[0094] Construct objective function: objective function = ∑ (observed value - theoretical value) 2 ;

[0095] The observed value is the target position information output by the trajectory generation unit, the theoretical value is the assumed fitting function, and the objective function is the loss function;

[0096] Obtain the fitting function that makes the objective function obtain the minimum value;

[0097] Form a formula to express the vehicle-following driving trajectory;

[0098]

[0099]

[0100] i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point;

[0101] j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight.

[0102] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. It will also be understood that, unless expressly defined herein, terms such as those defined in general dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not interpreted in an ideal or overly formal sense.

[0103] The present invention has been described in detail above through specific implementation modes and embodiments, but these do not constitute limitations of the present invention. Without departing from the principles of the present invention, those skilled in the art may also make many variations and improvements, which should also be regarded as the protection scope of the present invention.

Claims

1. A method for generating a preceding vehicle trajectory, characterized in that: The following steps are involved: S1, perceive the target and judge whether it is a new target or a historical target; S2: If it is a new target and there is a historical trajectory, the new target is integrated into its historical trajectory; if it is a new target and there is no historical trajectory, the new target specified trajectory is calculated by the position of the vehicle and the position of the preceding vehicle; If it is a historically valid target, the historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed; S3, by arranging trajectory points on the designated trajectory for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates, and subsequent trajectory fitting; S4, forming a following vehicle driving trajectory by fitting functions, including: S4.1, construct objective function: objective function = ∑ (observed value - theoretical value) 2 ; The observed value is the target position information obtained through steps S1 and S2, the theoretical value is the assumed fitting function, and the objective function is the loss function; S4.2, obtaining the fitting function corresponding to the minimum value of the objective function; S4.3, forming a formula to express the vehicle-following driving trajectory; i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point; j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight.

2. The method for generating a preceding vehicle trajectory according to claim 1, wherein: The specified trajectory is a parabola.

3. The method for generating a preceding vehicle trajectory according to claim 1, wherein: The trajectory points are arranged in an equally divided manner.

4. A preceding vehicle trajectory generation module, characterized in that: include: A receiving unit, wherein the self-sensing system receives target information, wherein the target information includes location information, index information and validity; Judgment unit, judging whether it is a new target or a historical target; The trajectory generation unit, if it is a new target and there is no historical trajectory, merges the new target into its historical trajectory; if it is a new target and there is no historical trajectory, calculates the new target specified trajectory through the position of the vehicle and the position of the preceding vehicle; If it is a historically valid target, the historically valid target information is updated to the information at the previous moment; if it is a historically invalid target, it is not processed; A fitting unit, which forms a following vehicle driving trajectory through a fitting function by arranging trajectory points for describing the historical trajectory of the preceding vehicle, subsequent trajectory updates and subsequent trajectory fitting on the designated trajectory; The fitting unit forms the following vehicle trajectory in the following way: Construct objective function: objective function = ∑ (observed value - theoretical value) 2 ; The observed value is the target position information output by the trajectory generation unit, the theoretical value is the assumed fitting function, and the objective function is the loss function; Obtain the fitting function that makes the objective function obtain the minimum value; Form a formula to express the vehicle-following driving trajectory; i: indicates the number of trajectory points; x i : represents the ordinate of the trajectory point; y i : represents the horizontal coordinate of the trajectory point; j: represents the order of the polynomial; ω j : represents the coefficient of the polynomial; λ: represents the weight.

5. The preceding vehicle trajectory generation module according to claim 4, characterized in that: The specified trajectory is a parabola.

6. The preceding vehicle trajectory generation module according to claim 4, characterized in that: The trajectory points are arranged in an equally divided manner.